1 / 66

CS184b: Computer Architecture (Abstractions and Optimizations)

Explore the requirements, topologies, and issues related to interconnect and routing in computer architecture. Learn about different routing styles and their impact on throughput and predictability.

Download Presentation

CS184b: Computer Architecture (Abstractions and Optimizations)

An Image/Link below is provided (as is) to download presentation Download Policy: Content on the Website is provided to you AS IS for your information and personal use and may not be sold / licensed / shared on other websites without getting consent from its author. Content is provided to you AS IS for your information and personal use only. Download presentation by click this link. While downloading, if for some reason you are not able to download a presentation, the publisher may have deleted the file from their server. During download, if you can't get a presentation, the file might be deleted by the publisher.

E N D

Presentation Transcript


  1. CS184b:Computer Architecture(Abstractions and Optimizations) Day 4: April 4, 2005 Interconnect

  2. Previously • CS184a • interconnect needs and requirements • basic topology • Mostly thought about static/offline routing

  3. This Quarter • This quarter • parallel systems require • typically dynamic switching • interfacing issues • model, hardware, software

  4. Today • Issues • Topology/locality/scaling • (some review) • Styles • from static • to online, packet, wormhole • Online routing

  5. Old Bandwidth aggregate, per endpoint local contention and hotspots Latency Cost (scaling) locality New Arbitration conflict resolution deadlock Routing (quality vs. complexity) Ordering (of messages) Issues

  6. Topology and Locality (Partially) Review

  7. $ $ $ $ P P P P Memory Simple Topologies: Bus • Single Bus • simple, cheap • low bandwidth • not scale with PEs • typically online arbitration • can be offline scheduled

  8. Offline: divide time into N slots assign positions to various communications run modulo N w/ each consumer/producer send/receiving on time slot e.g. 1: A->B 2: C->D 3: A->C 4: A->B 5: C->B 6: D->A 7: D->B 8: A->D Bus Routing

  9. Online: request bus wait for acknowledge Priority based: give to highest priority which requests consider ordering Goti = Wanti ^ Availi Availi+1=Availi ^ /Wanti Solve arbitration in log time using parallel prefix For fairness start priority at different node use cyclic parallel prefix deal with variable starting point Bus Routing

  10. want got avail Priority Arbitration Logic

  11. $ $ $ $ P P P P Memory Token Ring • On bus • delay of cycle goes as N • can’t avoid, even if talking to nearest neighbor • Token ring • pipeline bus data transit (ring) • high frequency • can exit early if local • use token to arbitrate use of bus

  12. $ $ $ P P P Multiple Busses • Simple way to increase bandwidth • use more than one bus • Can be static or dynamic assignment to busses • static • A->B always uses bus 0 • C->D always uses bus 1 • dynamic • arbitrate for a bus, like instruction dispatch to k identical CPU resources $ P

  13. Crossbar • No bandwidth reduction • (except receiver at endoint) • Easy routing (on or offline) • Scales poorly • N2 area and delay • No locality

  14. Hypercube • Arrange N=2n nodes in n-dimensional cube • At most n hops from source to sink • N = log2(N) • High bisection bandwidth • good for traffic (but can you use it?) • bad for cost [O(n2)] • Exploit locality • Node size grows • as log(N) [IO] • Maybe log2(N) [xbar between dimensions]

  15. Multistage • Unroll hypercube vertices so log(N), constant size switches per hypercube node • solve node growth problem • lose locality • similar good/bad points for rest

  16. Hypercube/Multistage Blocking • Minimum length multistage • many patterns cause bottlenecks • e.g.

  17. CS184a: Day16 Beneš Network • 2log2(N)-1 stages (switches in path) • Made of N/2 22 switchpoints [4 sw] • 4Nlog2(N) total switches • Compute route in O(N log(N)) time • Routes all permutations

  18. Online Hypercube Blocking • If routing offline, can calculate Benes-like route • Online, don’t have time, or global view • Observation: only a few, canonically bad patterns • Solution: Route to random intermediate • then route from there to destination • …turns worst-case into average case • at the expense of locality

  19. K-ary N-cube • Alternate reduction from hypercube • restrict to N<log(Nodes) dimensional structure • allow more than 2 ordinates in each dimension • E.g. mesh (2-cube), 3D-mesh (3-cube) • Matches with physical world structure • Bounds degree at node • Has Locality • Even more bottleneck potentials • make channels wider (CS184a:Day 17)

  20. Torus • Wrap around n-cube ends • 2-cube  cylinder • 3-cube  donut • Cuts worst-case distances in half • Can be laid-out reasonable efficiently • maybe 2x cost in channel width?

  21. Fat-Tree • Saw that communications typically has locality (CS184a) • Modeled recursive bisection/Rent’s Rule • Leiserson showed Fat-Tree was (area, volume) universal • w/in log(N) the area of any other structure • exploit physical space limitations wiring in {2,3}-dimensions

  22. MoT/Express Cube(Mesh with Bypass) • Large machine in 2 or 3 D mesh • routes must go through square/cube root switches • vs. log(N) in fat-tree, hypercube, MIN • Saw practically can go further than one hop on wire… • Add long-wire bypass paths

  23. Routing Styles

  24. Issues/Axes • Throughput of Communication relative to data rate of media • Single point-to-point link consume media BW? • Can share links between multiple comm streams? • What is the sharing factor? • Binding time/Predictability of Interconnect • Pre-fab • Before communication then use for long time • Cycle-by-cycle • Network latency vs. persistence of communication • Comm link persistence

  25. Axes Sharefactor (Media Rate/App. Rate) Persistence Predictability Net Latency

  26. Hardwired • Direct, fixed wire between two points • E.g. Conventional gate-array, std. cell • Efficient when: • know communication a priori • fixed or limited function systems • high load of fixed communication • often control in general-purpose systems • links carry high throughput traffic continually between fixed points

  27. Configurable • Offline, lock down persistent route. • E.g. FPGAs • Efficient when: • link carries high throughput traffic • (loaded usefully near capacity) • traffic patterns change • on timescale >> data transmission

  28. Time-Switched • Statically scheduled, wire/switch sharing • E.g. TDMA, NuMesh, TSFPGA • Efficient when: • thruput per channel < thruput capacity of wires and switches • traffic patterns change • on timescale >> data transmission

  29. Axes Time Mux Sharefactor (Media Rate/App. Rate) Predictability

  30. Self-Route, Circuit-Switched • Dynamic arbitration/allocation, lock down routes • E.g. METRO/RN1 • Efficient when: • instantaneous communication bandwidth is high (consume channel) • lifetime of comm. > delay through network • communication pattern unpredictable • rapid connection setup important

  31. Axes Phone; Videoconf; Cable Circuit Switch Sharefactor (Media Rate/App. Rate) Persistence Circuit Switch Net Latency Predictability

  32. Self-Route, Store-and-Forward, Packet Switched • Dynamic arbitration, packetized data • Get entire packet before sending to next node • E.g. nCube, early Internet routers • Efficient when: • lifetime of comm < delay through net • communication pattern unpredictable • can provide buffer/consumption guarantees • packets small

  33. Store-and-Forward

  34. Self-Route, Virtual Cut Through • Dynamic arbitration, packetized data • Start forwarding to next node as soon as have header • Don’t pay full latency of storing packet • Keep space to buffer entire packet if necessary • Efficient when: • lifetime of comm < delay through net • communication pattern unpredictable • can provide buffer/consumption guarantees • packets small

  35. Virtual Cut Through Three words from same packet

  36. Self-Route, Wormhole Packet-Switched • Dynamic arbitration, packetized data • E.g. Caltech MRC, Modern Internet Routers • Efficient when: • lifetime of comm < delay through net • communication pattern unpredictable • can provide buffer/consumption guarantees • message > buffer length • allow variable (? Long) sized messages

  37. Wormhole Single Packet spread through net when not stalled

  38. Wormhole Single Packet spread through net when stalled.

  39. Phone; Videoconf; Cable Circuit Switch Persistence Packet Switch IP Packet SMS message Axes Packet Switch Time Mux Sharefactor (Media Rate/App. Rate) Circuit Switch Config urable Predictability Net Latency

  40. Online Routing

  41. Time Mux Dynamic Costs: Area • Area • switch (1-1.5Kl2/ switch) • larger with pipeline (4Kl2) and rebuffer • state (SRAM bit = 1.2Kl2/ bit) • multiple in time-switched cases • arbitrartion/decision making • usually dominates above • buffering (SRAM cell per buffer) • can dominate

  42. Area (queue rough approx; you will refine)

  43. Costs: Latency • Time single path • make decisions • round-trip flow-control • Time contention/traffic • blocking in buffers • quality of decision • pick wrong path • have stale data

  44. Intermediate Approach • For large # of predictable patterns • switching memory may dominate allocation area • area of routed case < time-switched • [e.g. Large Cycles] • Get offline, global planning advantage • by source routing • source specifies offline determined route path • offline plan avoids contention

  45. Offline vs. Online • If know patterns in advance • offline cheaper • no arbitration (area, time) • no buffering • use more global data • better results • As becomes less predictable • benefit to online routing

  46. [example from Li and McKinley, IEEE Computer v26n2, 1993] Deadlock • Possible to introduce deadlock • Consider wormhole routed mesh

  47. Dimension Order Routing • Simple (early Caltech) solution • order dimensions • force complete routing in lower dimensions before route in next higher dimension

  48. [example from Li and McKinley, IEEE Computer v26n2, 1993] Dimension Ordered Routing • Route Y, then Route X

  49. Dimension Order Routing • Avoids cycles in channel graph • Limits routing freedom • Can cause artificial congestion • consider • (0,0) to (3,3) • (1,0) to (3,2) • (2,0) to (3,1) • [There is a rich literature on how to do better]

  50. Turn Model • Problem is cycles • Selectively disallow turns to break cycles • 2D Mesh West-First Routing

More Related